Advances in Applied Energy最新文献

Negative electrodes in vanadium redox flow batteries: recent advances and outlook 钒氧化还原液流电池负极研究进展及展望

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2026-01-08 DOI: 10.1016/j.adapen.2026.100262

Joel K. Edison , Adetunji Alabi , Ayoob Alhammadi , M. Shariq Anwar , Khalid Al-Ali , Sivaprakash Sengodan , Musbaudeen O. Bamgbopa , Giovanni Palmisano

Vanadium redox flow batteries are among the most promising technologies for large-scale energy storage, offering high safety, scalability, and stable performance through an aqueous electrolyte system. These attributes enable efficient integration of intermittent renewable sources such as wind and solar power. However, widespread deployment is constrained by limitations at the negative electrode, where the V³⁺/V²⁺ redox couple exhibits sluggish kinetics compared to the V⁵⁺/V⁴⁺ reaction at the positive electrode. Graphite felt, the most commonly used electrode material, satisfies several desirable properties: high conductivity, chemical stability, and large surface area; but suffers poor electrochemical activity toward V³⁺/V²⁺ conversion. This mismatch in reaction rates, combined with parasitic hydrogen evolution within the operating potential window and polarization losses, significantly reduces overall battery efficiency. This review critically examines strategies to overcome these challenges, including the role of functional groups on the electrode surface, electrocatalyst incorporation, and alternative electrode materials. The discussion begins with an analysis of fundamental limitations, side reactions, and degradation mechanisms, followed by criteria for electrode selection and the role of surface functional groups in enhancing kinetics. Approaches for electrocatalyst design and their impact on reaction rates are highlighted, and future research directions are proposed to accelerate the development of such batteries. By addressing the bottleneck at the negative electrode, these advancements aim to improve energy efficiency and durability, paving the way for broader adoption of such redox flow battery technology in grid-scale applications.

钒氧化还原液流电池是最有前途的大规模储能技术之一，通过水电解质系统提供高安全性，可扩展性和稳定的性能。这些特性使间歇性可再生能源（如风能和太阳能）能够有效整合。然而，由于负极的限制，广泛部署受到限制，其中V3+/V2+氧化还原对与正极的V5+/V4+反应相比，表现出缓慢的动力学。石墨毡是最常用的电极材料，满足几个理想的性能：高导电性，化学稳定性和大表面积；但对V3+/V2+转换的电化学活性较差。这种反应速率的不匹配，再加上工作电位窗口内的寄生氢释放和极化损耗，显著降低了电池的整体效率。这篇综述批判性地研究了克服这些挑战的策略，包括功能基团在电极表面的作用，电催化剂的掺入和替代电极材料。讨论从分析基本限制、副反应和降解机制开始，然后是电极选择的标准和表面官能团在增强动力学中的作用。重点介绍了电催化剂的设计方法及其对反应速率的影响，并提出了未来的研究方向，以加快此类电池的发展。通过解决负极的瓶颈，这些进步旨在提高能源效率和耐用性，为在电网规模应用中更广泛地采用这种氧化还原液流电池技术铺平道路。

{"title":"Negative electrodes in vanadium redox flow batteries: recent advances and outlook","authors":"Joel K. Edison , Adetunji Alabi , Ayoob Alhammadi , M. Shariq Anwar , Khalid Al-Ali , Sivaprakash Sengodan , Musbaudeen O. Bamgbopa , Giovanni Palmisano","doi":"10.1016/j.adapen.2026.100262","DOIUrl":"10.1016/j.adapen.2026.100262","url":null,"abstract":"<div><div>Vanadium redox flow batteries are among the most promising technologies for large-scale energy storage, offering high safety, scalability, and stable performance through an aqueous electrolyte system. These attributes enable efficient integration of intermittent renewable sources such as wind and solar power. However, widespread deployment is constrained by limitations at the negative electrode, where the V<sup>3+</sup>/V<sup>2+</sup> redox couple exhibits sluggish kinetics compared to the V<sup>5+</sup>/V<sup>4+</sup> reaction at the positive electrode. Graphite felt, the most commonly used electrode material, satisfies several desirable properties: high conductivity, chemical stability, and large surface area; but suffers poor electrochemical activity toward V<sup>3+</sup>/V<sup>2+</sup> conversion. This mismatch in reaction rates, combined with parasitic hydrogen evolution within the operating potential window and polarization losses, significantly reduces overall battery efficiency. This review critically examines strategies to overcome these challenges, including the role of functional groups on the electrode surface, electrocatalyst incorporation, and alternative electrode materials. The discussion begins with an analysis of fundamental limitations, side reactions, and degradation mechanisms, followed by criteria for electrode selection and the role of surface functional groups in enhancing kinetics. Approaches for electrocatalyst design and their impact on reaction rates are highlighted, and future research directions are proposed to accelerate the development of such batteries. By addressing the bottleneck at the negative electrode, these advancements aim to improve energy efficiency and durability, paving the way for broader adoption of such redox flow battery technology in grid-scale applications.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100262"},"PeriodicalIF":13.8,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146037900","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A RAG data pipeline transforming heterogeneous data into AI-ready format for autonomous building performance discovery RAG数据管道将异构数据转换为ai就绪格式，用于自主建筑性能发现

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-12-31 DOI: 10.1016/j.adapen.2025.100261

Han Li, Ana Comesana, Christopher Weyandt, Tianzhen Hong

Despite a growing volume of data being collected from buildings, a small portion of the data has been analyzed to provide actionable insights, mainly due to the labor intensive and error prone process of integrating and understanding the heterogeneous data with varying levels of quality and resolutions. This study introduces a novel domain-adapted Retrieval Augmented Generation (RAG) based data pipeline designed for efficient information retrieval and management from heterogeneous building data sources. The proposed data pipeline leverages Large Language Models (LLMs) and domain-specific processing to transform unstructured data (including images, tables, and plain text) into semantically rich, AI-ready representations. Evaluation results show that the domain-adapted processing techniques achieved approximately 30% improvement in both coverage and structural preservation for documents like images and tables compared to generic baseline methods employing generic prompts and minimal preprocessing. This automation capability transforms raw building artifacts into AI-ready searchable knowledge bases. A critical application demonstrated is the automation of semantic data model creation, reducing the manual effort from potentially days to minutes. This domain-adapted RAG pipeline significantly addresses persistent data challenges in the building sector, increasing the accessibility and utility of building information for diverse stakeholders and applications.

尽管从建筑物中收集的数据量越来越大，但只有一小部分数据被分析以提供可操作的见解，这主要是由于集成和理解具有不同质量和分辨率水平的异构数据的劳动密集型和容易出错的过程。本文提出了一种新的基于领域适应检索增强生成（RAG）的数据管道，用于从异构建筑数据源中高效地检索和管理信息。提议的数据管道利用大型语言模型（llm）和特定于领域的处理，将非结构化数据（包括图像、表和纯文本）转换为语义丰富的ai就绪表示。评估结果表明，与使用通用提示和最小预处理的通用基线方法相比，领域适应处理技术在图像和表格等文档的覆盖率和结构保存方面实现了大约30%的改进。这种自动化功能将原始的建筑工件转换为人工智能可搜索的知识库。演示的一个关键应用程序是语义数据模型创建的自动化，将人工工作从可能的几天减少到几分钟。这个领域适应的RAG管道显著地解决了建筑领域中持续存在的数据挑战，为不同的利益相关者和应用程序增加了建筑信息的可访问性和实用性。

{"title":"A RAG data pipeline transforming heterogeneous data into AI-ready format for autonomous building performance discovery","authors":"Han Li, Ana Comesana, Christopher Weyandt, Tianzhen Hong","doi":"10.1016/j.adapen.2025.100261","DOIUrl":"10.1016/j.adapen.2025.100261","url":null,"abstract":"<div><div>Despite a growing volume of data being collected from buildings, a small portion of the data has been analyzed to provide actionable insights, mainly due to the labor intensive and error prone process of integrating and understanding the heterogeneous data with varying levels of quality and resolutions. This study introduces a novel domain-adapted Retrieval Augmented Generation (RAG) based data pipeline designed for efficient information retrieval and management from heterogeneous building data sources. The proposed data pipeline leverages Large Language Models (LLMs) and domain-specific processing to transform unstructured data (including images, tables, and plain text) into semantically rich, AI-ready representations. Evaluation results show that the domain-adapted processing techniques achieved approximately 30% improvement in both coverage and structural preservation for documents like images and tables compared to generic baseline methods employing generic prompts and minimal preprocessing. This automation capability transforms raw building artifacts into AI-ready searchable knowledge bases. A critical application demonstrated is the automation of semantic data model creation, reducing the manual effort from potentially days to minutes. This domain-adapted RAG pipeline significantly addresses persistent data challenges in the building sector, increasing the accessibility and utility of building information for diverse stakeholders and applications.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100261"},"PeriodicalIF":13.8,"publicationDate":"2025-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145939211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Provincial CO2 emission peak and drivers in China based on near-real-time data 基于近实时数据的中国各省二氧化碳排放峰值及驱动因素分析

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-12-29 DOI: 10.1016/j.adapen.2025.100260

Xiaoting Huang , Zhu Deng , Nanzhuo Jin , Duo Cui , Zhu Liu

As effective multilevel climate governance involves integrated action and reporting at and across national and subnational levels, tracking mitigation performance requires more timely emission monitoring. To overcome the lag of current emission inventories, here we propose a subnational near-real-time CO₂ emission accounting framework and integrate our daily time series (2019 - present) with existing annual inventories to construct a consistent 2000 - 2024 emissions time series for 30 provinces (excluding Tibet, Hong Kong, Macao, and Taiwan due to data availability). Based on this dataset, we further develop a moving-window Mann-Kendall trend test algorithm to detect the peak status and confirm its robustness through sensitivity tests, which systematically characterizes provincial emission trajectories and classifies provinces into three peaking-status categories. The results show that by 2024, two provinces (Beijing and Jilin) are in a post-peak decline phase, 15 provinces are still rising, and the remaining 13 provinces exhibit a volatile or uncertain peak phase. The provinces that have peaked or are fluctuating form a northeast-southwest belt, whereas most provinces in the southeast and northwest are on a rising trajectory. The attribution analysis reveals that, although provinces approaching a plateau are increasingly driven by energy structure shifts, emissions decline in several northeastern and industrial provinces is linked to population loss or economic slowdown, indicating the risks of passive rather than structural decarbonization. Overall, our near-real-time diagnostic framework provides a transferable approach for continuous monitoring of subnational emissions stages, offering timely evidence for China’s progress toward peaking and informing broader high-frequency monitoring, reporting, and verification (MRV) efforts.

由于有效的多层次气候治理涉及国家和国家以下各级的综合行动和报告，跟踪缓解绩效需要更及时地监测排放。为了克服当前排放清单的滞后性，本文提出了一个次国家级近实时二氧化碳排放核算框架，并将我们的每日时间序列（2019年至今）与现有的年度清单相结合，构建了30个省份（由于数据可用性，不包括西藏、香港、澳门和台湾）2000 - 2024年的一致排放时间序列。在此基础上，进一步开发了移动窗口Mann-Kendall趋势检验算法，通过灵敏度检验验证了算法的鲁棒性，系统地表征了各省的排放轨迹，并将各省划分为三种峰值状态。结果表明，到2024年，北京和吉林2省处于峰后下降阶段，15省仍在上升，其余13省呈现波动或不确定的峰值阶段。达到峰值或波动的省份形成东北-西南带，而东南部和西北部的大多数省份处于上升轨道。归因分析表明，虽然接近平台的省份越来越多地受到能源结构转变的驱动，但东北部分省份和工业省份的排放量下降与人口流失或经济放缓有关，表明被动脱碳的风险大于结构性脱碳的风险。总体而言，我们的近实时诊断框架为持续监测次国家级排放阶段提供了一种可转移的方法，为中国达到峰值的进展提供了及时的证据，并为更广泛的高频监测、报告和核查（MRV）工作提供了信息。

{"title":"Provincial CO2 emission peak and drivers in China based on near-real-time data","authors":"Xiaoting Huang , Zhu Deng , Nanzhuo Jin , Duo Cui , Zhu Liu","doi":"10.1016/j.adapen.2025.100260","DOIUrl":"10.1016/j.adapen.2025.100260","url":null,"abstract":"<div><div>As effective multilevel climate governance involves integrated action and reporting at and across national and subnational levels, tracking mitigation performance requires more timely emission monitoring. To overcome the lag of current emission inventories, here we propose a subnational near-real-time CO<sub>2</sub> emission accounting framework and integrate our daily time series (2019 - present) with existing annual inventories to construct a consistent 2000 - 2024 emissions time series for 30 provinces (excluding Tibet, Hong Kong, Macao, and Taiwan due to data availability). Based on this dataset, we further develop a moving-window Mann-Kendall trend test algorithm to detect the peak status and confirm its robustness through sensitivity tests, which systematically characterizes provincial emission trajectories and classifies provinces into three peaking-status categories. The results show that by 2024, two provinces (Beijing and Jilin) are in a post-peak decline phase, 15 provinces are still rising, and the remaining 13 provinces exhibit a volatile or uncertain peak phase. The provinces that have peaked or are fluctuating form a northeast-southwest belt, whereas most provinces in the southeast and northwest are on a rising trajectory. The attribution analysis reveals that, although provinces approaching a plateau are increasingly driven by energy structure shifts, emissions decline in several northeastern and industrial provinces is linked to population loss or economic slowdown, indicating the risks of passive rather than structural decarbonization. Overall, our near-real-time diagnostic framework provides a transferable approach for continuous monitoring of subnational emissions stages, offering timely evidence for China’s progress toward peaking and informing broader high-frequency monitoring, reporting, and verification (MRV) efforts.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100260"},"PeriodicalIF":13.8,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145939210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Residential vehicle-to-grid profits under dynamic pricing: The role of real-world charging behavior 动态定价下住宅车辆入网利润：现实世界充电行为的作用

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-12-29 DOI: 10.1016/j.adapen.2025.100259

Sonia Martin , William A. Paxton , Ram Rajagopal

Vehicle-to-grid (V2G) charging is a promising grid asset that is currently hampered by concerns of high capital costs and battery degradation. Simulations of V2G charging with passenger electric vehicles (EVs) have demonstrated profitability under static electricity prices and ancillary electricity market participation. However, it is unknown whether V2G is economically viable with participation in nascent dynamic pricing and demand response (DR) programs available in 2025. Furthermore, it has not been determined which EV drivers will financially benefit from these programs. In a case study of California’s Pacific Gas & Electric utility territory, we simulate annual and net present costs over a vehicle’s lifetime with V2G participation in dynamic pricing and DR programs, testing sensitivity to charger access, installation costs, and battery degradation. With high charger access, favorable dynamic pricing, and low installation costs, V2G can yield a profit of up to $1,181 per vehicle per year with a 4.5 year break-even time for the charger and installation costs. However, if these criteria are not met, the system may take 5–15 years to pay itself off. Leveraging historical EV data to illustrate behavior-based differences in V2G profits, these findings suggest that V2G adoption should be limited to specific groups of drivers with certain dynamic pricing structures, unless installation costs decrease in the future grid.

车辆到电网（V2G）充电是一项很有前途的电网资产，但目前受到高资本成本和电池退化问题的阻碍。在静态电价和辅助电力市场参与的情况下，乘用电动汽车V2G充电的仿真证明了其盈利能力。然而，目前尚不清楚V2G在2025年参与新兴的动态定价和需求响应（DR）计划时是否具有经济可行性。此外，目前还没有确定哪些电动汽车司机将从这些计划中获得经济利益。在加州太平洋天然气和电力公用事业领域的案例研究中，我们模拟了V2G参与动态定价和DR计划的车辆生命周期的年度和净现值成本，测试了对充电器接入、安装成本和电池退化的敏感性。V2G具有较高的充电器使用率、优惠的动态定价和较低的安装成本，每辆车每年可产生高达1181美元的利润，充电器和安装成本的盈亏平衡时间为4.5年。然而，如果不满足这些标准，该系统可能需要5-15年才能收回成本。利用历史电动汽车数据来说明基于行为的V2G利润差异，这些发现表明，除非未来电网的安装成本降低，否则V2G的采用应限于具有特定动态定价结构的特定驾驶员群体。

{"title":"Residential vehicle-to-grid profits under dynamic pricing: The role of real-world charging behavior","authors":"Sonia Martin , William A. Paxton , Ram Rajagopal","doi":"10.1016/j.adapen.2025.100259","DOIUrl":"10.1016/j.adapen.2025.100259","url":null,"abstract":"<div><div>Vehicle-to-grid (V2G) charging is a promising grid asset that is currently hampered by concerns of high capital costs and battery degradation. Simulations of V2G charging with passenger electric vehicles (EVs) have demonstrated profitability under static electricity prices and ancillary electricity market participation. However, it is unknown whether V2G is economically viable with participation in nascent dynamic pricing and demand response (DR) programs available in 2025. Furthermore, it has not been determined which EV drivers will financially benefit from these programs. In a case study of California’s Pacific Gas & Electric utility territory, we simulate annual and net present costs over a vehicle’s lifetime with V2G participation in dynamic pricing and DR programs, testing sensitivity to charger access, installation costs, and battery degradation. With high charger access, favorable dynamic pricing, and low installation costs, V2G can yield a profit of up to $1,181 per vehicle per year with a 4.5 year break-even time for the charger and installation costs. However, if these criteria are not met, the system may take 5–15 years to pay itself off. Leveraging historical EV data to illustrate behavior-based differences in V2G profits, these findings suggest that V2G adoption should be limited to specific groups of drivers with certain dynamic pricing structures, unless installation costs decrease in the future grid.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100259"},"PeriodicalIF":13.8,"publicationDate":"2025-12-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145939209","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Planning future charging infrastructure for private EVs: A city-scale assessment of demand and capacity 规划未来私人电动汽车的充电基础设施：城市规模的需求和容量评估

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-12-07 DOI: 10.1016/j.adapen.2025.100258

Hong Yuan , Minda Ma , Nan Zhou , Yanqiao Deng , Junhong Liu , Shufan Zhang , Zhili Ma

Capacity planning for electric vehicle (EV) charging infrastructure has emerged as a critical challenge in developing low-carbon urban energy systems. This study proposes the first demand-driven, multi-objective planning model for optimizing city-scale capacity allocation of EV charging infrastructure. The model employs a bottom-up approach to estimate charging demand differentiated by vehicle type—battery electric vehicles (BEVs), extended-range electric vehicles (EREVs), and plug-in hybrid electric vehicles (PHEVs). Chongqing, a rapidly expanding EV industry cluster in China with a strong industrial base, supportive policies, and diverse urban morphologies, is selected as the case study. The results show that (1) monthly EV electricity consumption in Chongqing rose from 18.9 gigawatt-hours (GWh) in June 2022 to 57.5 GWh in December 2024, with associated carbon emissions increasing from 9.9 kilotons of carbon dioxide (ktCO₂) to 30 ktCO₂; (2) 181,622 additional charging piles were installed between 2022 and 2024, with the fastest growth observed in Yubei, reflecting a demand-responsive strategy that prioritizes areas with higher population density, higher income levels, and adequate land availability for pile deployment, rather than broad geographic coverage; and (3) between 2025 and 2030, EV electricity demand is projected to reach 1940 GWh, with the number of charging piles exceeding 1.4 million, and charging demand from EREVs and PHEVs expected to overtake BEVs later in the period. While Chongqing serves as the pilot area, the proposed planning platform is adaptable for application in cities worldwide, enabling cross-regional comparisons under diverse socio-economic, geographic, and policy conditions. Overall, this work offers policymakers a versatile tool to support sustainable, cost-effective EV infrastructure deployment aligned with low-carbon electrification targets in the transportation sector.

电动汽车（EV）充电基础设施的容量规划已成为发展低碳城市能源系统的关键挑战。本文首次提出了城市规模电动汽车充电基础设施容量配置优化的需求驱动多目标规划模型。该模型采用自下而上的方法来估计不同车型（纯电动汽车（bev）、增里程电动汽车（EREVs）和插电式混合动力汽车（phev））的充电需求。重庆是中国快速发展的电动汽车产业集群，拥有强大的产业基础、扶持政策和多样的城市形态。结果表明：(1)重庆市电动汽车月用电量从2022年6月的18.9 GWh增加到2024年12月的57.5 GWh，碳排放量从9.9千吨二氧化碳（ktCO2）增加到30千吨二氧化碳；(2) 2022年至2024年期间，新增充电桩181,622个，其中渝北增长最快，反映了需求响应战略，优先考虑人口密度高、收入水平高、土地可用性充足的地区，而不是广泛的地理覆盖；(3) 2025 - 2030年，电动汽车电力需求预计将达到1940 GWh，充电桩数量将超过140万个，纯电动汽车和插电式混合动力汽车的充电需求预计将在后期超过纯电动汽车。虽然重庆是试点地区，但拟议的规划平台可适用于全球城市，可以在不同的社会经济、地理和政策条件下进行跨区域比较。总的来说，这项工作为政策制定者提供了一个多功能工具，以支持可持续的、具有成本效益的电动汽车基础设施部署，并与交通部门的低碳电气化目标保持一致。

{"title":"Planning future charging infrastructure for private EVs: A city-scale assessment of demand and capacity","authors":"Hong Yuan , Minda Ma , Nan Zhou , Yanqiao Deng , Junhong Liu , Shufan Zhang , Zhili Ma","doi":"10.1016/j.adapen.2025.100258","DOIUrl":"10.1016/j.adapen.2025.100258","url":null,"abstract":"<div><div>Capacity planning for electric vehicle (EV) charging infrastructure has emerged as a critical challenge in developing low-carbon urban energy systems. This study proposes the first demand-driven, multi-objective planning model for optimizing city-scale capacity allocation of EV charging infrastructure. The model employs a bottom-up approach to estimate charging demand differentiated by vehicle type—battery electric vehicles (BEVs), extended-range electric vehicles (EREVs), and plug-in hybrid electric vehicles (PHEVs). Chongqing, a rapidly expanding EV industry cluster in China with a strong industrial base, supportive policies, and diverse urban morphologies, is selected as the case study. The results show that (1) monthly EV electricity consumption in Chongqing rose from 18.9 gigawatt-hours (GWh) in June 2022 to 57.5 GWh in December 2024, with associated carbon emissions increasing from 9.9 kilotons of carbon dioxide (ktCO<sub>2</sub>) to 30 ktCO<sub>2</sub>; (2) 181,622 additional charging piles were installed between 2022 and 2024, with the fastest growth observed in Yubei, reflecting a demand-responsive strategy that prioritizes areas with higher population density, higher income levels, and adequate land availability for pile deployment, rather than broad geographic coverage; and (3) between 2025 and 2030, EV electricity demand is projected to reach 1940 GWh, with the number of charging piles exceeding 1.4 million, and charging demand from EREVs and PHEVs expected to overtake BEVs later in the period. While Chongqing serves as the pilot area, the proposed planning platform is adaptable for application in cities worldwide, enabling cross-regional comparisons under diverse socio-economic, geographic, and policy conditions. Overall, this work offers policymakers a versatile tool to support sustainable, cost-effective EV infrastructure deployment aligned with low-carbon electrification targets in the transportation sector.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100258"},"PeriodicalIF":13.8,"publicationDate":"2025-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145749721","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Economics of electric vehicle corridor fast charging in the United States 美国电动汽车走廊快速充电的经济性

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-12-06 DOI: 10.1016/j.adapen.2025.100257

Brennan Borlaug , Vince Caristo , Fletcher Ouren , Fan Yang , Eric Wood , Laura Roberson

Corridor direct-current fast charging (DCFC) stations enable long-distance electric vehicle travel, yet their economics remain uncertain due to high capital costs, low initial utilization, and exposure to utility demand charges. This study evaluates the long-term economics of corridor DCFC across the United States, incorporating capital and operating expenses—including charging equipment and real-world utility tariffs—alongside modeled station utilization, financial incentives, and ancillary retail revenue. In the Baseline scenario, modeled breakeven costs for corridor DCFC average $0.42/kWh over 20 years, yet fewer than half of stations reach cost parity with gasoline on a per-mile basis. Utilization is the primary driver of cost variation, with low-utilization stations costing roughly six times more per kilowatt-hour than the national average. Excluding stations that fail to reach cost parity reduces National Highway System coverage within 50 miles from 94% to 67%, underscoring the trade-off between market-driven deployment and comprehensive network coverage. These results provide guidance for charging providers, utilities, planners, and policymakers seeking to develop and sustain a financially viable national corridor charging network.

走廊直流电快速充电站（DCFC）可以实现电动汽车的长途行驶，但由于资本成本高、初始利用率低以及公用事业需求收费，其经济性仍不确定。本研究评估了美国走廊DCFC的长期经济效益，将资本和运营费用（包括充电设备和现实世界的公用事业费率）与模拟站点利用率、财政激励和辅助零售收入结合起来。在基线情景中，走廊dfc的模型盈亏平衡成本在20年内平均为0.42美元/千瓦时，但不到一半的加油站达到每英里成本与汽油相同。利用率是成本变化的主要驱动因素，低利用率电站每千瓦时的成本大约是全国平均水平的六倍。排除那些无法达到成本平价的站点，将使50英里范围内的国家公路系统覆盖率从94%降至67%，强调了市场驱动部署和全面网络覆盖之间的权衡。这些结果为寻求发展和维持财政上可行的国家走廊充电网络的充电供应商、公用事业公司、规划人员和政策制定者提供了指导。

{"title":"Economics of electric vehicle corridor fast charging in the United States","authors":"Brennan Borlaug , Vince Caristo , Fletcher Ouren , Fan Yang , Eric Wood , Laura Roberson","doi":"10.1016/j.adapen.2025.100257","DOIUrl":"10.1016/j.adapen.2025.100257","url":null,"abstract":"<div><div>Corridor direct-current fast charging (DCFC) stations enable long-distance electric vehicle travel, yet their economics remain uncertain due to high capital costs, low initial utilization, and exposure to utility demand charges. This study evaluates the long-term economics of corridor DCFC across the United States, incorporating capital and operating expenses—including charging equipment and real-world utility tariffs—alongside modeled station utilization, financial incentives, and ancillary retail revenue. In the Baseline scenario, modeled breakeven costs for corridor DCFC average $0.42/kWh over 20 years, yet fewer than half of stations reach cost parity with gasoline on a per-mile basis. Utilization is the primary driver of cost variation, with low-utilization stations costing roughly six times more per kilowatt-hour than the national average. Excluding stations that fail to reach cost parity reduces National Highway System coverage within 50 miles from 94% to 67%, underscoring the trade-off between market-driven deployment and comprehensive network coverage. These results provide guidance for charging providers, utilities, planners, and policymakers seeking to develop and sustain a financially viable national corridor charging network.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"21 ","pages":"Article 100257"},"PeriodicalIF":13.8,"publicationDate":"2025-12-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145705694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Advancing building stock transformation models: An agent-based approach and its application to Germany 推进存量转换模型：基于主体的方法及其在德国的应用

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-11-20 DOI: 10.1016/j.adapen.2025.100256

Şirin Alibaş , Songmin Yu , Mahsa Bagheri , Tobias Fleiter

The building sector is pivotal for achieving climate neutrality, requiring sophisticated modeling tools to guide the energy transition. It is highly heterogeneous with variety in both the built environment and in its decision-makers. While many sectoral models exist, most of them fall short of capturing the barriers and limitations in the energy transition as they lack high spatial resolution and a detailed representation of heterogeneous building characteristics and local infrastructure constraints. To address this gap, we present RENDER-Building, a new agent-based model (ABM) designed for high-resolution analysis of building stock transformation. We validate and apply the model to the German building stock to simulate potential transformation pathways until 2050 under three distinct scenarios. The individual buildings are the agents here with detailed attributes, located in a settlement type in a NUTS3 region. The model explicitly considers the availability of energy infrastructure and simulates agents’ decisions about renovation and technology adoption based on bounded rationality. Our case study’s results indicate that even with ambitious measures, Germany’s building sector may miss its short-term emission targets due to the inertia of the existing stock. A transformation pathway considering realistic challenges could substantially exceed the short- and long-term emission targets, necessitating difficult and potentially costly interventions to get back on track. Our study demonstrates the utility of high-resolution ABMs in providing nuanced, actionable insights for policymakers, helping them to navigate the complexities of the building sector’s energy transition.

建筑行业是实现气候中和的关键，需要复杂的建模工具来指导能源转型。它在建筑环境和决策者方面都是高度异质的。虽然存在许多部门模型，但大多数模型都无法捕捉能源转型中的障碍和限制，因为它们缺乏高空间分辨率和对异质建筑特征和当地基础设施限制的详细表示。为了解决这一差距，我们提出了一种新的基于代理的模型（ABM），用于建筑存量转换的高分辨率分析。我们验证了该模型并将其应用于德国建筑存量，以模拟在三种不同情景下到2050年的潜在转型路径。单个建筑是具有详细属性的代理，位于NUTS3区域的聚落类型中。该模型明确地考虑了能源基础设施的可用性，并基于有限理性模拟了主体关于改造和技术采用的决策。我们的案例研究结果表明，即使采取雄心勃勃的措施，由于现有库存的惯性，德国建筑行业也可能无法实现其短期排放目标。考虑到现实挑战的转型途径可能大大超过短期和长期排放目标，需要采取困难和可能昂贵的干预措施才能回到正轨。我们的研究展示了高分辨率ABMs在为政策制定者提供细致入微、可操作的见解方面的效用，帮助他们应对建筑行业能源转型的复杂性。

{"title":"Advancing building stock transformation models: An agent-based approach and its application to Germany","authors":"Şirin Alibaş , Songmin Yu , Mahsa Bagheri , Tobias Fleiter","doi":"10.1016/j.adapen.2025.100256","DOIUrl":"10.1016/j.adapen.2025.100256","url":null,"abstract":"<div><div>The building sector is pivotal for achieving climate neutrality, requiring sophisticated modeling tools to guide the energy transition. It is highly heterogeneous with variety in both the built environment and in its decision-makers. While many sectoral models exist, most of them fall short of capturing the barriers and limitations in the energy transition as they lack high spatial resolution and a detailed representation of heterogeneous building characteristics and local infrastructure constraints. To address this gap, we present RENDER-Building, a new agent-based model (ABM) designed for high-resolution analysis of building stock transformation. We validate and apply the model to the German building stock to simulate potential transformation pathways until 2050 under three distinct scenarios. The individual buildings are the agents here with detailed attributes, located in a settlement type in a NUTS3 region. The model explicitly considers the availability of energy infrastructure and simulates agents’ decisions about renovation and technology adoption based on bounded rationality. Our case study’s results indicate that even with ambitious measures, Germany’s building sector may miss its short-term emission targets due to the inertia of the existing stock. A transformation pathway considering realistic challenges could substantially exceed the short- and long-term emission targets, necessitating difficult and potentially costly interventions to get back on track. Our study demonstrates the utility of high-resolution ABMs in providing nuanced, actionable insights for policymakers, helping them to navigate the complexities of the building sector’s energy transition.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"20 ","pages":"Article 100256"},"PeriodicalIF":13.8,"publicationDate":"2025-11-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145568635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A systematic review of technologies, measures, and CO2 emission reduction potential for maritime transport decarbonisation 对海上运输脱碳的技术、措施和二氧化碳减排潜力进行系统审查

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-11-07 DOI: 10.1016/j.adapen.2025.100255

Sina Fadaie , Patricia Thornley , Jean-Baptiste Souppez

The maritime shipping sector is a significant contributor to global carbon dioxide (CO₂) emissions, accounting for approximately 2.7%-3% of global emissions. In response, the International Maritime Organization (IMO) has set ambitious targets: a 30% reduction in emissions by 2030, 80% by 2040, and net-zero by 2050, relative to 2008 levels. Meeting these goals requires a comprehensive understanding of the full range of viable decarbonisation measures. Therefore, this study conducts a systematic review of maritime decarbonisation measures, applying the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Unlike previous studies, this paper not only provides an updated overview of CO₂ reduction measures but also maps them to specific vessel types based on data reported in the literature. Furthermore, the findings are compared with literature to highlight shifts in mitigation potential. A case study is also included to schematically demonstrate how these measures can be applied in practice. Following a rigorous analysis: (i) thirty-two individual CO₂ mitigation measures were identified and classified into six categories, (ii) alternative fuels shown the highest long-term potential (5–100 % CO₂ emission reduction), whereas hull design improvements show the lowest (1–20 %), (iii) the wide disparity in reported abatement values is attributed to inconsistent system boundaries, variability in fuel origin, partial-blend scenarios, and differing assumptions across studies, (iv) combinations of measures provide the most practical and realistic pathway to phased emissions reduction. These findings are expected to assist decision-makers in selecting effective, context-appropriate strategies to support global maritime decarbonisation and ensure long-term sectoral sustainability.

海运部门是全球二氧化碳（CO2）排放的重要贡献者，约占全球排放量的2.7%-3%。为此，国际海事组织（IMO）制定了雄心勃勃的目标：到2030年减排30%，到2040年减排80%，到2050年实现净零排放。要实现这些目标，就需要全面了解各种可行的脱碳措施。因此，本研究对海事脱碳措施进行了系统审查，应用了系统审查和荟萃分析（PRISMA）方法的首选报告项目。与以往的研究不同，本文不仅提供了二氧化碳减排措施的最新概述，而且根据文献中报道的数据将其映射到特定的船舶类型。此外，将研究结果与文献进行比较，以突出减缓潜力的变化。还包括一个案例研究，以示意性地演示如何在实践中应用这些措施。经过严格的分析：(i)确定了32种单独的二氧化碳减缓措施，并将其分为6类；（ii）替代燃料显示出最高的长期潜力（5 - 100%的二氧化碳减排），而船体设计改进显示出最低的长期潜力（1 - 20%）；（iii）报告的减排值的巨大差异归因于不一致的系统边界、燃料来源的可变性、部分混合情景和不同研究的不同假设。（iv）各项措施的结合为分阶段减少排放提供了最实际和最现实的途径。这些研究结果有望帮助决策者选择有效的、适合具体情况的战略，以支持全球海上脱碳，并确保长期的部门可持续性。

{"title":"A systematic review of technologies, measures, and CO2 emission reduction potential for maritime transport decarbonisation","authors":"Sina Fadaie , Patricia Thornley , Jean-Baptiste Souppez","doi":"10.1016/j.adapen.2025.100255","DOIUrl":"10.1016/j.adapen.2025.100255","url":null,"abstract":"<div><div>The maritime shipping sector is a significant contributor to global carbon dioxide (CO<sub>2</sub>) emissions, accounting for approximately 2.7%-3% of global emissions. In response, the International Maritime Organization (IMO) has set ambitious targets: a 30% reduction in emissions by 2030, 80% by 2040, and net-zero by 2050, relative to 2008 levels. Meeting these goals requires a comprehensive understanding of the full range of viable decarbonisation measures. Therefore, this study conducts a systematic review of maritime decarbonisation measures, applying the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Unlike previous studies, this paper not only provides an updated overview of CO<sub>2</sub> reduction measures but also maps them to specific vessel types based on data reported in the literature. Furthermore, the findings are compared with literature to highlight shifts in mitigation potential. A case study is also included to schematically demonstrate how these measures can be applied in practice. Following a rigorous analysis: (i) thirty-two individual CO<sub>2</sub> mitigation measures were identified and classified into six categories, (ii) alternative fuels shown the highest long-term potential (5–100 % CO<sub>2</sub> emission reduction), whereas hull design improvements show the lowest (1–20 %), (iii) the wide disparity in reported abatement values is attributed to inconsistent system boundaries, variability in fuel origin, partial-blend scenarios, and differing assumptions across studies, (iv) combinations of measures provide the most practical and realistic pathway to phased emissions reduction. These findings are expected to assist decision-makers in selecting effective, context-appropriate strategies to support global maritime decarbonisation and ensure long-term sectoral sustainability.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"20 ","pages":"Article 100255"},"PeriodicalIF":13.8,"publicationDate":"2025-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145519708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Utilizing oxygen from green hydrogen production in wastewater treatment plant aeration: A techno-economic analysis 利用废水处理厂曝气中绿色制氢产生的氧气：技术经济分析

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-11-01 DOI: 10.1016/j.adapen.2025.100254

Levin Matz , Lukas Koenemann , Steffen Brundiers , Inga Beyers , Arne Freyschmidt , Astrid Bensmann , Richard Hanke-Rauschenbach

The growing demand for green hydrogen is driving the expansion of water electrolysis. The resulting oxygen byproduct offers potential added value when used in sectors with high oxygen demand, such as wastewater treatment. This study investigates the techno-economic viability of using electrolysis oxygen to supplement conventional air blowers in the aeration process of municipal wastewater treatment plants (WWTPs) to reduce aeration costs and thereby improve the overall economics of hydrogen production. A comprehensive system model is developed, incorporating renewable electricity supply, water electrolysis, hydrogen compression, storage, and transport, as well as WWTP aeration via conventional air blowers and electrolysis oxygen. Results show that electrolysis oxygen can reduce WWTP aeration costs by up to 68%. If these cost reductions are attributed as a benefit to the hydrogen system, they correspond to hydrogen supply cost savings of up to 0.39 EUR/kgH₂. However, the analysis indicates that economic viability is substantially influenced by factors such as the distance of hydrogen transport from the WWTP to the European Hydrogen Backbone feed-in point, which should not exceed 25 km, and the alignment between the scale of hydrogen production and the size of the WWTP, with cost-effective integration being particularly feasible for larger WWTPs (≥500,000 PE).

对绿色氢日益增长的需求正在推动水电解的扩张。当用于废水处理等高需氧量行业时，所产生的氧气副产品具有潜在的附加值。本研究探讨了在城市污水处理厂（WWTPs）曝气过程中使用电解氧作为常规鼓风机的补充，以降低曝气成本，从而提高制氢的整体经济可行性。开发了一个综合系统模型，包括可再生电力供应、水电解、氢气压缩、储存和运输，以及通过传统鼓风机和电解氧气进行污水处理厂曝气。结果表明，电解氧可使污水处理厂曝气成本降低68%。如果将这些成本的降低归因于氢气系统的效益，则相当于节省了高达0.39欧元/千瓦时的氢气供应成本。然而，分析表明，经济可行性在很大程度上受到诸如从污水处理厂到欧洲氢气骨干进料点的氢气输送距离（不应超过25公里）以及氢气生产规模与污水处理厂规模之间的一致性等因素的影响，对于较大的污水处理厂（≥500,000 PE）而言，成本效益整合尤其可行。

{"title":"Utilizing oxygen from green hydrogen production in wastewater treatment plant aeration: A techno-economic analysis","authors":"Levin Matz , Lukas Koenemann , Steffen Brundiers , Inga Beyers , Arne Freyschmidt , Astrid Bensmann , Richard Hanke-Rauschenbach","doi":"10.1016/j.adapen.2025.100254","DOIUrl":"10.1016/j.adapen.2025.100254","url":null,"abstract":"<div><div>The growing demand for green hydrogen is driving the expansion of water electrolysis. The resulting oxygen byproduct offers potential added value when used in sectors with high oxygen demand, such as wastewater treatment. This study investigates the techno-economic viability of using electrolysis oxygen to supplement conventional air blowers in the aeration process of municipal wastewater treatment plants (WWTPs) to reduce aeration costs and thereby improve the overall economics of hydrogen production. A comprehensive system model is developed, incorporating renewable electricity supply, water electrolysis, hydrogen compression, storage, and transport, as well as WWTP aeration via conventional air blowers and electrolysis oxygen. Results show that electrolysis oxygen can reduce WWTP aeration costs by up to 68%. If these cost reductions are attributed as a benefit to the hydrogen system, they correspond to hydrogen supply cost savings of up to 0.39 EUR/kgH<sub>2</sub>. However, the analysis indicates that economic viability is substantially influenced by factors such as the distance of hydrogen transport from the WWTP to the European Hydrogen Backbone feed-in point, which should not exceed 25 km, and the alignment between the scale of hydrogen production and the size of the WWTP, with cost-effective integration being particularly feasible for larger WWTPs (≥500,000 PE).</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"20 ","pages":"Article 100254"},"PeriodicalIF":13.8,"publicationDate":"2025-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145466306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multimodal ultra-short-term probabilistic solar power forecasting with generative AI and Transformer 基于生成式人工智能和变压器的多模态超短期概率太阳能发电预测

IF 13.8 Q1 ENERGY & FUELS

Advances in Applied Energy

Pub Date : 2025-10-31 DOI: 10.1016/j.adapen.2025.100250

Binyu Xiong , Yuntian Chen , Xin Zhao , Zhongyue Su , Jun Fu , Dali Chen , Dongxiao Zhang

Solar power is one of the most widely adopted forms of renewable energy, with its usage rapidly increasing in recent years. However, solar power generation is highly sensitive to environmental factors, leading to frequent fluctuations. These fluctuations present challenges for large-scale integration into the power grid. Sky images, which provide real-time information about sky conditions, can help improve ultra-short-term solar power forecasting. Nonetheless, due to the inherent uncertainty of cloud movement, solar power generation can be unstable over short time intervals. This instability calls for the inclusion of uncertainty in ultra-short-term forecasting models. Inspired by recent advances in deep learning, we propose a novel framework for ultra-short-term probabilistic solar power generation forecasting. This framework uses both historical solar power generation data and sky images as inputs. First, a video prediction model based on generative artificial intelligence (AI) is applied to generate multiple potential future sky image sequences. Next, a Transformer-based multimodal model combines each predicted sequence with historical solar power generation data to derive a distribution of possible future power outputs. To account for uncertainties in both the video prediction and multimodal models, these distributions are aggregated into a single overall distribution. We conducted experiments on real-world datasets, with prediction times ranging from 15 to 60 min, to compare our method with previous mainstream approaches. The results demonstrate that our method outperforms existing approaches in both deterministic and probabilistic forecasting tasks. In the 15 min ahead forecasting task, compared to the method using only historical solar power generation data, our method reduces the Root Mean Square Error (RMSE) of the deterministic evaluation metric by 20.6% and the Continuous Ranked Probability Score (CRPS) of the probabilistic evaluation metric by 19.4%. When compared to the method using only sky image data, our approach reduces the RMSE by 47.3% and the CRPS by 51.3%. Furthermore, we conduct additional analysis on the performance of various methods under different weather conditions.

太阳能是最广泛采用的可再生能源之一，近年来其使用量迅速增加。然而，太阳能发电对环境因素高度敏感，导致波动频繁。这些波动对大规模并入电网提出了挑战。天空图像提供了关于天空状况的实时信息，可以帮助改进超短期太阳能预测。然而，由于云层运动的固有不确定性，太阳能发电在短时间间隔内可能不稳定。这种不稳定性要求在超短期预测模型中包含不确定性。受深度学习最新进展的启发，我们提出了一个超短期概率太阳能发电预测的新框架。该框架使用历史太阳能发电数据和天空图像作为输入。首先，应用基于生成式人工智能（AI）的视频预测模型生成多个潜在的未来天空图像序列。接下来，基于变压器的多模态模型将每个预测序列与历史太阳能发电数据结合起来，得出可能的未来功率输出分布。为了考虑视频预测和多模态模型中的不确定性，这些分布被汇总成一个整体分布。我们在真实世界的数据集上进行了实验，预测时间从15到60分钟不等，将我们的方法与之前的主流方法进行了比较。结果表明，我们的方法在确定性和概率预测任务中都优于现有的方法。在提前15分钟预测任务中，与仅使用历史太阳能发电数据的方法相比，我们的方法将确定性评价指标的均方根误差（RMSE）降低了20.6%，将概率评价指标的连续排序概率得分（CRPS）降低了19.4%。与仅使用天空图像数据的方法相比，我们的方法将RMSE降低了47.3%，CRPS降低了51.3%。此外，我们还对各种方法在不同天气条件下的性能进行了额外分析。

{"title":"Multimodal ultra-short-term probabilistic solar power forecasting with generative AI and Transformer","authors":"Binyu Xiong , Yuntian Chen , Xin Zhao , Zhongyue Su , Jun Fu , Dali Chen , Dongxiao Zhang","doi":"10.1016/j.adapen.2025.100250","DOIUrl":"10.1016/j.adapen.2025.100250","url":null,"abstract":"<div><div>Solar power is one of the most widely adopted forms of renewable energy, with its usage rapidly increasing in recent years. However, solar power generation is highly sensitive to environmental factors, leading to frequent fluctuations. These fluctuations present challenges for large-scale integration into the power grid. Sky images, which provide real-time information about sky conditions, can help improve ultra-short-term solar power forecasting. Nonetheless, due to the inherent uncertainty of cloud movement, solar power generation can be unstable over short time intervals. This instability calls for the inclusion of uncertainty in ultra-short-term forecasting models. Inspired by recent advances in deep learning, we propose a novel framework for ultra-short-term probabilistic solar power generation forecasting. This framework uses both historical solar power generation data and sky images as inputs. First, a video prediction model based on generative artificial intelligence (AI) is applied to generate multiple potential future sky image sequences. Next, a Transformer-based multimodal model combines each predicted sequence with historical solar power generation data to derive a distribution of possible future power outputs. To account for uncertainties in both the video prediction and multimodal models, these distributions are aggregated into a single overall distribution. We conducted experiments on real-world datasets, with prediction times ranging from 15 to 60 min, to compare our method with previous mainstream approaches. The results demonstrate that our method outperforms existing approaches in both deterministic and probabilistic forecasting tasks. In the 15 min ahead forecasting task, compared to the method using only historical solar power generation data, our method reduces the Root Mean Square Error (RMSE) of the deterministic evaluation metric by 20.6% and the Continuous Ranked Probability Score (CRPS) of the probabilistic evaluation metric by 19.4%. When compared to the method using only sky image data, our approach reduces the RMSE by 47.3% and the CRPS by 51.3%. Furthermore, we conduct additional analysis on the performance of various methods under different weather conditions.</div></div>","PeriodicalId":34615,"journal":{"name":"Advances in Applied Energy","volume":"20 ","pages":"Article 100250"},"PeriodicalIF":13.8,"publicationDate":"2025-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145466307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0